Need alone is an incomplete guide
Severe problems can still be weak places to give. Where proven interventions are scarce, or few organizations can deliver them well, additional funding is unlikely to go far.
Every engagement moves from a funder's intent to a cost-estimated recommendation, with a different question and framework at each stage.
What impact does the funder want to create, and for whom?
We work with the funder to turn their goals into a clear research question, agreeing who they want to reach, where, at what scale, and what level of impact they're aiming for.
We look at how important, neglected and solvable each problem is, a lens drawn from effective giving, to find where each additional rupee can do the most good.
Where is the need greatest, where can philanthropy make the most difference, and what works?
We map the need at a local level, look at what the government already spends, and bring together the evidence on which interventions work.
RIO’s “NLIO” lens: need, leverage, interventions and the organisations delivering them
An evidence hierarchy adapted from Cochrane, J-PAL and 3ie
Public data from NFHS, NITI Aayog, RBI, and central and state budgets
Which intervention does the most good per rupee?
We build cost-effectiveness models for the shortlisted interventions, adjusted for local conditions and tested against a range of assumptions.
Cost-effectiveness analysis adapted from GiveWell and Founders Pledge
Which organizations can deliver this well, and put the funding to good use?
We identify organizations via our own search or an open call, screen for eligibility and evidence, and follow up with field visits and expert input.
We look at fit with the cause area, the capacity to absorb new funding, and basic compliance. We also assess the strength of each organization's own evidence and how faithfully it delivers its programmes.
Did our advice change where the money went?
Once the funder commits, we ask what they would have funded without our advice.
Funders split their commitment across three scenarios: it would have reached equally effective organizations anyway, gone to the cause without an evidence lens, or gone unspent. We then adjust for how confident they are in their answers.
Throughout, we test desk findings against people working in the states we study, because district statistics conceal much of what happens on the ground.
Severe problems can still be weak places to give. Where proven interventions are scarce, or few organizations can deliver them well, additional funding is unlikely to go far.
Many strong Indian organizations have invested in programs ahead of formal evaluation. So we draw on the evidence behind the intervention and assess how well each organization delivers it.
While every cause area we consider matters, our choice reflects where we believe our research can add the most right now. We assessed six cause areas against four questions:
for smallholder farmers
The two selected areas combine an enduring problem, existing funding to build on, real policy salience, and a sector mature enough that additional money has somewhere productive to go. They also sit at opposite ends of the evidence spectrum: early childhood has a deep global evidence base that transfers imperfectly, while smallholder incomes required evidence to be assembled largely from the ground up.
For many interventions in India the upper rungs are empty, because the rigorous trials were run elsewhere. We triangulate strong international evidence with Indian public data, from sources such as NITI Aayog and the Reserve Bank of India, and with implementers' own monitoring data. The discipline lies in taking the best evidence seriously wherever it was produced, while being honest that an effect measured in one setting seldom transfers cleanly to another.
We express cost-effectiveness as a ratio of benefit to cost (BCR), which is a calculation of the benefit generated from an intervention at a given cost. A multiple of 10x means our model estimates that a programme creates ten times the value of the cost of the intervention given as a direct transfer. In livelihood work, value is the additional income a household earns over time; in early childhood, it is the gain in a child's development and earnings potential.
Programmes that appear similar can differ greatly in what they achieve per rupee. In early childhood, 60% of funding flows to interventions with an estimated 4–5x return, while options returning 15–80x remain underfunded. For a funder, where the money goes matters as much as how much is given.
Every estimate rests on four inputs. We set out each one, with its source, for any funder who wants to check it.
BCR=(Effect × Internal validity × External validity) ÷ Cost=value per dollar
Small changes in assumptions move the result, so every estimate carries a range.
20x and 18x are effectively the same answer; 20x and 2x are not.
Comparing a child's development with a farmer's income requires a judgement of values, which we leave to the funder.
They guide a decision before the money moves; they do not replace evaluation afterwards.
In our research on early childhood, we identified ~30 interventions and Kangaroo Mother Care (KMC) was one of them. Kangaroo Mother Care, skin to skin contact for underweight newborns, is one of the most evidence-backed, low-cost interventions in newborn survival, recommended by WHO and linked to significant reductions in mortality and infection.
Trials claim Kangaroo Mother Care reduces newborn mortality by 40%. When evaluating KMC for Maharashtra, Raising Impact adjusted for local baseline practices, real-world scaling drop-offs, and bundled care attribution—refining the estimate down to a realistic ~14%. We bridge the gap between evidence and operational reality so funders can deploy capital with confidence.
The widely cited 40% reduction in newborn mortality became a working estimate of ~14% for Maharashtra.
A global effect size is a starting point and the work lies in adjusting it for local conditions, one stated assumption at a time.
If you’re a funder planning to make a significant philanthropic commitment, we would be glad to talk it through with you.
Write to info@raisingimpact.org